From: Andrej Bauer <andrej.bauer@andrej.com>
To: David Leduc <david.leduc6@googlemail.com>
Cc: categories@mta.ca
Subject: Re: Equality as an adjunction
Date: Fri, 3 Sep 2010 00:54:39 +0200 [thread overview]
Message-ID: <E1OrfeA-0007e2-3m@mlist.mta.ca> (raw)
In-Reply-To: <E1Or9Uc-00034j-JW@mlist.mta.ca>
On Thu, Sep 2, 2010 at 3:40 AM, David Leduc <david.leduc6@googlemail.com> wrote:
> He writes the definition as a bidirectional typing rule for the
> internal language of a suitable category.
>
> Phi |- P(x,x) [x:A]
> ===================
> Phi, x=x' |- P(x,x') [x x':A]
>
> What are the left and right adjoint functors here?
Short answer: the right adjoint is contraction, the left adjoint is
(making conjunction with) equality.
Long answer:
To make things easier to understand, let me interpret types as sets,
predicates as subsets and entailment |- a the subset relation. For a
set X let Sub(X) be the powerset of X ordered by subset inclusion. Fix
a set A and define functors (monotone maps between posets)
G : Sub(A x A) --> Sub(A)
F : Sub(A) --> Sub(A x A)
by
G(P) = {x \in A | P(x,x)}
F(Q) = {(x,y) \in A x A | Q(x) and x = y}
Thus G is composition with the diagonal map (a.k.a. contraction in
logic) and F is intersection with the diagonal (equality relation).
These two functors are adjoint, since for any P in Sub(A x A) and Q in
Sub(A) we have
Q \susbeteq G(P) <==> F(Q) \susbeteq P
If we write this in logical form and plug in definitions of F and G we get:
Q(x) |- P(x,x) [x : A]
==================================
Q(x), x=x' |- P(x,x') [x,x' : A]
which is more or less what you wrote.
With kind regards,
Andrej
[For admin and other information see: http://www.mta.ca/~cat-dist/ ]
next prev parent reply other threads:[~2010-09-02 22:54 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-09-02 1:40 David Leduc
2010-09-02 15:22 ` Robert Seely
2010-09-02 17:56 ` Michael Shulman
2010-09-02 22:54 ` Andrej Bauer [this message]
2010-09-03 7:06 ` Vaughan Pratt
2010-09-04 4:03 ` David Leduc
[not found] ` <AANLkTikUtD3SbB+4OBRpniqgLRBg0szKYkMSwiWx+ycr@mail.gmail.com>
2010-09-04 7:21 ` David Leduc
[not found] ` <Pine.LNX.4.64.1009041105300.2080@prism.math.mcgill.ca>
2010-09-05 2:26 ` David Leduc
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=E1OrfeA-0007e2-3m@mlist.mta.ca \
--to=andrej.bauer@andrej.com \
--cc=categories@mta.ca \
--cc=david.leduc6@googlemail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).